arXiv 2510.212072025-10-24cs.LG

自适应图残差专家混合：异构专业化下的多样图无监督学习

Adaptive Graph Mixture of Residual Experts: Unsupervised Learning on Diverse Graphs with Heterogeneous Specialization

School of New Media and Communication, Tianjin University（新媒体与传播学院，天津大学）；Baidu（百度）；School of Computer Science and Engineering, Beihang University（计算机科学与工程学院，北航大学）

Yunlong Chu, Minglai Shao, Zengyi Wo, Bing Hao, Yuhang Liu, Ruijie Wang, Jianxin Li

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

提出ADaMoRE框架，通过骨干-残差专家架构和结构感知门控网络，在无监督条件下实现异构图专家混合训练，提升数据效率与训练稳定性，在16个基准上达到最先进性能。

摘要：

图神经网络（GNNs）面临一个根本性的适应性挑战：其固定的消息传递架构难以应对现实世界图数据的巨大多样性，其中最优计算策略因局部结构和任务而异。尽管混合专家（MoE）为适应性提供了一条有前景的路径，但现有的图MoE方法仍受限于对监督信号的依赖以及训练异构专家时的不稳定性。我们提出了ADaMoRE（自适应残差专家混合），一个原则性框架，能够在图上实现鲁棒的、完全无监督的异构MoE训练。ADaMoRE采用骨干-残差专家架构，其中基础编码器提供稳定性，而专门的残差专家捕获多样化的计算模式。一个结构感知的门控网络执行细粒度的节点路由。整个架构使用统一的非监督目标进行端到端训练，该目标整合了主要重建任务与信息论多样性正则化器，以显式地在专家之间强制执行功能专业化。理论分析证实了我们的设计提高了数据效率和训练稳定性。在16个基准上的广泛评估验证了ADaMoRE在无监督节点分类和少样本学习中的最先进性能，以及在多样图数据和任务上优越的泛化能力、训练效率和更快的收敛速度。

英文摘要：

Graph Neural Networks (GNNs) face a fundamental adaptability challenge: their fixed message-passing architectures struggle with the immense diversity of real-world graphs, where optimal computational strategies vary by local structure and task. While Mixture-of-Experts (MoE) offers a promising pathway to adaptability, existing graph MoE methods remain constrained by their reliance on supervised signals and instability when training heterogeneous experts. We introduce ADaMoRE (Adaptive Mixture of Residual Experts), a principled framework that enables robust, fully unsupervised training of heterogeneous MoE on graphs. ADaMoRE employs a backbone-residual expert architecture where foundational encoders provide stability while specialized residual experts capture diverse computational patterns. A structurally-aware gating network performs fine-grained node routing. The entire architecture is trained end-to-end using a unified unsupervised objective, which integrates a primary reconstruction task with an information-theoretic diversity regularizer to explicitly enforce functional specialization among the experts. Theoretical analysis confirms our design improves data efficiency and training stability. Extensive evaluation across 16 benchmarks validates ADaMoRE's state-of-the-art performance in unsupervised node classification and few-shot learning, alongside superior generalization, training efficiency, and faster convergence on diverse graphs and tasks.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。